Brain Death: Rapid Evaluation Using Computerised Radionucleide Cerebral Flow Study

Med. J.Malaysia Vol. 41 No. 3 September 1986

BRAIN DEATH: RAPID EVALUATION USING

COMPUTERIZED RADIONUCLIDE CEREBRAL

FLOW STUDY

M. PARAMSOTHY GRACIE

s.

T. G. LOH A. E. DELlLKAN

SUMMARY

Demonstration of arrested intracerebral blood flow is the ultimate evidence of brain death. Computerized radionuclide cerebral flow study was done on 18 patients diagnosed clinically as brain dead. Correlation was made with clinical neurophysiological and EEG findings. The criteria for diagnosis of arrested intracerebral perfusion using radionuclide flow study were: non-visualiza-tion of blood flow activity in the intracranial arteries during the arterial phase, diffused cerebral activity during the capillary phase and non-filling of venous sinuses during the venous phase; visuali-zation of typical 'hot nasal' activity; the

time-M. Paramsothy, MRCP (UK) Gracie S. Y. Ong, FFARACS B. H. Wong, FFARACS

T. G. Loh, MRCP (London, Glasgow, Edin)

A. E. Delilkan, FFARCS, FFARACS

Departments of Radiology, Anaesthesiology and Medicine

Faculty of Medicine, University of Malaya 59100 Kuala Lumpur, Malaysia

For correspondence: Or M. Paramsothy Associate Professor

Division of Nuclear Medicine Radiology Unit, University Hospital 59100 Kuala Lumpur, Malaysia

activity curve over the cerebral hemispheres lacks a bolus effect and instead shows a delayed gradual rise of activity. These features are pathognomonic of brain tamponade.

Arrested intracranial circulation was seen in 16 patients (ten had electrocerebral silence; one had extremely abnormal EEG with small voltage activity and five had no EEG done). In the remain-ing two patients, some cerebral blood flow was

demonstrated (one had no definite cerebral

activity and the other had diffused EEG activity).

Radionuclide cerebral flow study is a very sensi-tive, accurate, safe, simple, rapid and non-invasive modality in confirming braindeath and is especially useful in patients on "brain-protection" regime, in hypothermia or in certain metabolic states where diagnosis based on clinical and EEG criteria is difficult. EEG need not be a required procedure once brain death is established by the demonstra-tion of arrested intracranialcirculademonstra-tion.

INTRODUCTION

leads to autolysis and necrosis and the brain at autopsy has a softened and liquefied appea-rance.2, 3 These changes do not occur under other circumstances of death.

When brain death has occurred further artificial support is futile and should be withdrawn for the following reasons: to reduce "suffering" of the patient, and to spare relatives from further emotional trauma and loss of finance; reduce wasteful drain of limited resources and burden on the hospital bed and the staff which could be used for other patients who may still benefit from them; make available the organs of these patients in the best possible condition for transplantation. Good quality cadaver kidneys are obtainable only in cases where cerebral death occurs before cardiac standstill and the surgeon removes the organs while they are still beingoxvqenated."

Unconsciousness, absent spontaneous respira-tion when ventilator is disconnected, absent spontaneous movement (decerebrate or epileptic), non-reacting dilated pupils, absent cephalic reflexes and electrocerebral silence at least 24 hours apart in the absence of hypothermia and depressant drugs are widely accepted as diagnostic of cerebral death." These criteria are open to criticisms as patients may present all signs of irreversible coma including electrocerebral silence and yet actually be in a reversible state. The extensive use of barbiturates, phenytoin and other anaesthetics as "brain protection" or to reduce intracranial hypertension7 makes the diagnosis of brain death based on clinical and EEG criteria only more difficult. It is now widely accepted that EEG is not necessary for diagnosis of brain death.8

, 9

In a few countries, an additional parameter using selective common carotid and vertebral angio-graphy is necessary by law to confirm the absence of intracranial circulation before proceeding with organ removal for transplantation.2,10,11

Radionuclide cerebral flow study using a portable scintillation probe or gamma cameral 2-1 5

can rapidly and reliably confirm absent intra-cranial circulation at bedside in brain dead patients. Recently intravenous and intra-arterial digital

subtraction cerebral anrl neck angiography (DSA) have been used to confirm absent intracranial circulation in the diagnosis of brain death.1

6 ,17

This technique is invasive, fraught with danger of contrast idiosyncracy, albeit low, and cannot be used at the bedside.

This paper describes our experience at the University Hospital, Kuala Lumpur, with radio-nuclide cerebral flow study (CFS) in the diagnosis of brain death. We find this technique an accurate, sensitive, simple, safe and rapid test to confirm absent intracerebral circulation in clinically diagnosed brain dead patients.

MATERIAL AND METHODS

Eighteen patients with clinical diagnosis of brain death with and without foreknowledge of EEG findings were studied. These patients were admitted to the Intensive Care Unit (ICU) of University Hospital, Kuala Lumpur between 3 February 1981 and 15 March 1986. They were managed on a standard "brain protection" or brain resuscitation regime which included intra-venous analgesia and anaesthetics and muscle relaxants; controlled ventilation maintaining adequate arterial oxygenation (P02

>

and moderate hypocarbia (PC02 3.5 to 4.5 kPa);

normotension. and avoidance of cerebral venous congestion; fluid retention and avoidance of salt-, free infusates; mannitol therapy.

This regime was maintained for 48 to 72 hours. Four hours or longer following cessation of central deafferentation agents and reversal of neuro-muscular block, neurological examination was performed and repeated 24 hours later.

The clinical diagnosis of brain death was made using the criteria issued by the conference of Medical Royal Colleges and their Faculties in the United Kingdom, 1976.8 Electroencephalo-graphy was performed in 13 patients using standards set out by the American Electroence-phalographic Society's "Ad Hoc Committee on EEG Criteria for the Determination of Cerebral Death" 18 In the remaining five patients, EEG

or patient's death. Autopsy was not done due to socio-religious constraints.

The patients included nine males and nine females aged 2 to 60, with a mean age of 20 years.

The causes of brain death (Table I) were: massive brain injury from traffic accidents, fall and assault; meningitis, encephalitis, and meningo-encephalitis; multiple brain secondaries; cardiac arrest from septicaemic shock, post-cardiac surgery, tight mitral stenosis in pregnancy, and anaesthetic induction; asphyxia with respiratory arrest from status asthmaticus and acute epiglot-titis; and metabolic encephalopathy in Reye's syndrome and diabetic hypoglycemic coma with disseminated intravascular coagulation (DIVC) and stroke. Respiratory failure and impaired circulation resulting in hypoxia and ischaemia of the brain underlied all these conditions.

TABLE I

CAUSE OF BRAIN DEATH IN 18 PATIENTS

Cause

Massive brain injury: motor veh icle accident fall

assault Brain infection:

meningitis encephalitis meningoencephalitis Brain tumour:

mu Itiple secondaries Cardiac arrest:

septicaemic shock post-cardiac surgery

tight mitral stenosis in pregnancy

diabetic gangrene during anaesthetic induction Asphyxia with respiratory arrest:

status asthmatics acute epiglottitis Metabolic encephalopathy:

Reye's syndrome

diabetic hypoglycemic coma with stroke and DiVe

Total

No.

2

1

1

1 1

1

1

2 1

2

18

With manual artificial ventilation maintained, the patient's head and neck were positioned beneath the face of a gamma-scintillation camera with parallel hole, low energy collimation and interfaced with a dedicated computer. 0.1 mCi/kg body weight of Technetium-99m pertechnetate was injected as a bolus with saline flush into an antecubital vein or viacentral venous pressure line. Sequential data in frame mode of tracer blood flow in the neck and head in anterior projection were acquired at 2 seconds interval for 60 seconds in 64 x 64 matrix format and stored in a magnetic disc. Simultaneous analog 2 sec x 30 sequential images were also obtained on X-ray films. The whole procedure took about five minutes. Five patients in coma due to various causes but not clinically brain dead, and five conscious and neurologically normal patients were studied in a similar manner to serve as controls. The digital images were examined and displayed both in static and cine fashion. Time activity (T/A) curves of regions of interest (ROI) over the cerebral hemisphere and carotid artery were created. The bolus was checked and if spread or fragmented, a repeat injection was made and data acquired in a similar manner. The curves were 9 point smoothed, using 1 :6:1 filter. Background sub-traction was not done.

Correlation was made with clinical and EEG findings and the ultimate fate of the patient and the interval of death from time of cerebral flow study. These patients when confirmed to have brain tamponade were continued on artifi-cial support but transferred from ICU to an ordinary ward.

RESULTS

2,!5 SEC /DIU

(e)eL.-_~...-.,:-=,"=--::,::,-- ~ VIP>

VIP> U I P>

UIP>FIG 18

Cl

never approached that of the usual peak and there was no definite drip in activity during the one-minute period of imaging. This gradual rise of T/ A curve is due to radioactivity from extra-cerebral (scalp) circulation. The T/A curves of the carotid vessels showed normal bol us pattern confirming a good tracer bolus injection. These features are diagnostic of arrested intracerebral circulation.

Fig. 2 Normal cerebral T/A curve (Cl) clearly shows bolus effect of intracranial flow. This characteristic sharp rise to peak and fall is indistinguishable from curves obtained on comatose but not brain dead patients.

Diffuse homogenous symmetrical and equal radioactivity in the distribution of the middle cerebral arteries in both cerebral hemispheres during the capillary phase and prompt emptying into the venous sinuses (S) during venous phase is clearly shown. T/A curve over the cerebral hemispheres in this patient showed the typical pattern with a sharp rise to peak and fall of activity as the bulk of the tracer bolus passed through the brain (Fig. 2). The coma patients without cl inical evidence of brain death showed comparable flow images and T/ A curves.

In 16 of the 18 clinically suspected brain dead patients, radionuclide flow study showed the neck vessels and scalp circulation to be normal, but there was no visual ization of the cerebral arteries and venous sinuses (Fig. 3). The typical "hot-nasal" activity was seen, probably due to hold-up of blood flow in the distal internal carotid vessels at the level of the circle of Willis and diver-sion of flow to facial bones (Fig. 3). The T/A curves in these patients (Fig. 4) were dramati-cally different from the normal flow pattern (Fig. 2) in that they showed on bolus effect. There was marked impairment in time and rate of appearance of radioactivity, the level of activity

'[~'f

...-:;:.r ••.•

',,11• _'~r

: "

'.:-" .

Normal radionuclide cerebral angiography. During the arterial phase flow radioactivity in the carotid arteries in neck (Cl, middle cerebral arteries (M) which is equal on both sides and in anterior cere-bral arteries (A) is visualized. The arterial-capillary phase show diffuse homogenous activity through-out the cerebral hemisphere. Activity in the saggital sinus (SI and region of jugular bulb is visualized during the venous phase.

11' .f

•

....

~. "

.. . .

. ;..'.

\.;

"'.);

C2

,0

:

<.

;~ '-N

.)' ' ; '

..

....

<6a:

bjp~ Ql ORl;~"I . " _

VIP> UIP>FIG le

Fig. 3 Flow images in brain death show during arterial phase activity in neck vessels (C) and scalp circu-lation but no visualization of intracranial circula-tion. The venous phase show no filling of the saggital sinus. The typical 'hot-nasal' activity (N) is seen.

(A)

.~.

vI P>

UIP>FIG lA

843 MAX

with no definite cerebral activity; the other patient had small (1.5 mm) nonreactive pupils

but otherwise nonreactive cephal ic reflexes and EEG done twice within 24 hours showed diffuse delta activity, some theta and occasional beta activity characterizing diffuse encephalopathy. Both these patients had cardiac standstill three days later. The majority of patients (72%) had cardiac standstill within one day of radionuclide study.

[2

DISCUSSION

The T/A curves over regions of the cerebral hemisphere represent passage of radioactive bolus through two different types of circulation: the large volume of blood flow through the relatively small cerebral blood vessels and cause a

charac-teristic bolus effect with a sharp rise and fall of radioactivity (Fig. 2); the relatively slower and dispersed extracerebral (scalp) peripheral blood flow fed by branches of the external carotid arteries does not show a bolus effect but a gradual rise of radioactivity (Fig. 4).

The criteria for diagnosis of arrested intra-cerebral perfusion in radionuclide cerebral flow study are: non-visualization of blood flow activity in the intracranial arteries during the arterial phase, diffuse cerebral activity during the capillary phase and non-filling of venous sinuses during the venous phase; visualization of typical "hot nasal" activity; and the time-activity curve over the cerebral hemispheres lacks a bolus effect and instead shows a delayed gradual rise of activity. These features are characteristic and pathogno-monic of brain tamponade.

2 5 SEC DIU

(0

V I P>

V I P)

VIP>

VIP>FIG 10

TABLEII

EEG AND RADIONUCLlDE CEREBRAL ANGIOGRAPHIC FINDINGS

Fig. 4 Cerebral T/ A curve (C2) in brain death shows no bolus effect. Th is delay and gradual rise of activity is from the relatively slower and dispersed extra-cerebral scalp peripheral type of circulation. EEG in this patient was iso-electric.

In two patients clinically suspected to have brain death, some intracerebral perfusion was demonstrated: one patient had absent cephal ic reflexes but intact spinal reflexes and an EEG Of these 16 patients with absent intracerebral perfusion demonstrated by the radionucl ide angiography, ten had isoelectric EEG (Tables 11,

Ill). The EEG in another patient done one day prior to radionuclide study was reported as extremely abnormal but not isoelectric. The remaining five patients did not have EEG done due to technical or time constraints or patients' death.

EEG silent EEG abnormal EEG not done

Total

Flow Present

o

2

Flow Absent

10

5

16

Total

11 2 5

18

It is important to look for filling of venous sinuses or gradual washout of cerebral radio-activity as in severely impairedIJut not totally

TABLE III

CLINICAL, EEG AND RADIONUCLlDE CFS FINDINGS

Patient _{Cause of} Interval in

Neurological EEG Radionuclide days from scan

brain injury examination intracerebral to cardiac

Age (yrsl

flow standstill

2 F MVA with fractured skull brain death isoelectric x 1 no flow and brain contusion

4 M Fall, extradural haematoma brain death isoelectric x 2 no flow same day cerebral oedem a and infarct

4 M Encephalitis with brain death isoelectric no flow same day

respiratory arrest pupils fixed, dilated next day

reflex plantar equivocal

14 M Skull fracture with brain death not done no flow

subdural haematoma spinal reflexes present

20 F Choriocarcinoma with brain brain death not done no flow secondaries and

intra-cerebral haemorrhage

4 M Nephrotic syndrome, brain death not done no flow same day

uraemia, septicaemic shock, meningitis

39 M MVA with fractured skull, brain death lsoelectric x 1 no flow same day CT scan: cerebral

contusion and oedema Respiratory arrest

31 F Pelvic abscess, septi- brain death isoelectric x 1 no flow caemic shock, cardiac

arrest

43 M Status asthmaticus, small (1.5mm), diffuse delta some flow in 3

asphyxia, cardiac non-reactive pupils, activ itv ove r anterior and

arrest otherwise cephalic both cerebral middle cerebral

reflexes non-reactive, hemispheres. arteries, saggital brain death Some theta and sinus visualized

occasional beta activity x 2

38 M Diabetic hypo- brain death isoelectric x 1 no flow same day

glycemia coma DIVC

8 M Reye 's syndrome, brain death isoelectric x 2 no flow

hepatic encephalopathy, plantar cardiac arre st

4 F Reye's syndrome absent cephalic not done no flow

Patient Interval in

Causes of Neurological EEG Badionuclide days from scan

Age brain injury examination intracerabraI to cardiac

(vrs) Sex flow standstll

3 M Tetralogy of Fallot, cerebral de ath isoelectric x 1 minimal tracer 3 Post-card iac surgery Spinal reflexes present flow in anterior

shock. Cardiac arrest and middle cerebral

vessels next day

2 F Acute epiglottitis, brain death isoelectric x 1 no flow 2 asphyxia, respiratory

arrest

27 F Mitral stenosis with brain death isoelectric x 1 no flow 5 pregnancy, sudden coma,

cardiac arrest

33 F Statu s asthmatics, brain death isoelectric x 1 no flow 6 asphyxia, respiratory,

arrest

22 F Meninqoencephalitis brain death basically flat no flow same day except for faint next day

occip ital small voltage activity

60 F Diabetic gangrene for brain death not done no flow 1 day

amputation. Cardiac arrest du ri n g anaesthe tic induction

to make the diagnosis; the analog flow images alone for all our patients were sufficient.

Radionuclide cerebral flow study is very sensitive and provides clinically useful estimation of gross changes in cerebral blood flow. Failure to show intracerebral vasculatu re may be compared to a non-filling cerebral contrast angiogram. The technique is simple, results are displayed within minutes and arterial punctures are not necessary. Depressant drugs used as 'brain protection' and hypothermia should have no effect on the results. The radiation exposure to the patient and staff is minimal and less than that would be acquired during a portable chest radiograph.

Where definitive aetiology and degree of brain damage is known decision on brain death can be based on a clinical basis alone and EEG is not

essential.8, 9 However, EEG is useful when the

nature of brain inju ry is not known provided there is absence of eNS depressant drugs, hypo-thermia and certain metabolic states." In their paper, Mohandas et. al.,9 reported instances of

and reversibility of brain damage.1 9 The presence of modest amounts of artefactual low voltage rhythmic or quasi-sinusoidal activity (usually of greater amplitude in non-cerebral leads) prevented the diagnosis of electrocerebral silence in patients whose brains were found to be totally necrotic at the time of EEG.2 o In our study one patient with absent cephalic reflexes had a basically flat EEG except for faint occipital small voltage activity but an arrested intracranial circulation was demon-strated by radionuclide flow study on the next day. It appears, therefore, in the presence of definite clinical-neurophysiological criteria of brain death and demonstration of arrested intra-cerebral perfusion by radionuclide flow study the value of an EEG examination is extremely questionable.

"The doctor's dilemma about brain death arises only when patients are put on ventilators, and it is therefore of his own making. When a patient who already has severe brain damage develops respiratory insufficiency or cardio-respiratory arrest, careful thought should be taken before artificial respiration is extended beyond the immediate resuscitation period. Prolongation of such a patient's life, even for 12 hours (it may be much longer in practice). reflects no credit on his doctors, particularlv if th is is done only so as to postpone the decision to let events take their natural course. It would be unfortunate if the time came when no patient in hospital can decently die without the last rite of modern medicine a statutory period on the ventilator."?1

CONCLUSION

Radionuclide cerebral flow study is a very sensitive, accurate, safe and rapid method to demonstrate arrested intracranial circulation in brain dead patients. It is a more practical and use-ful method than EEG to confirm brain death especially in busy intensive care units such as ours in the University Hospital, Kuala Lumpur where the pressure for space is acute (daily bed-occupancy is greater than 90%). The long duration for EEG examinations (including a repeat EEG in 24 hours); and the occurrence of artefactual voltage activity which prevents diagnosis of

electrocerebral silence in EEGs can prolong patients' stay adding burden on ICU bed and the staff which could be used for other patients who may still benefit from them. However where definite aetiology and degree of brain damage are known decision on brain death can be based on clinical basis alone and a second modality such as EEG or radionuclide cerebral flow study is not essential. The doctor's dilemma about with-drawal of further artificial respiratory support is often his own making. In patients with severe brain damage who develop respiratory in-sufficiency or cardio-respiratory arrest, careful thought should be taken before the decision to provide artificial support is made.

REFERENCES

Lindegaard K F, Grip A, Nordnes H. Precerebral haemodynamics in brain tamponade. Part I: Clinical studies on blood flow velocity.Neurochirurgia 1980; 23: 133-143.

2 Report of the Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death. A Definition of Irreversible Coma. JAMA 1968; 205 : 85-88.

3 Kramer W, Twynman J A. Acute intracranial hyper-tension; an experimental investigation. Brain Res 1976;6: 686-705.

4 Forrester A C. Brain death and the don ation of cadaver kidney s.Health Bulletin 1976; 34 : 199.

5 Walker A E, Brickford R, Aung Met. al. An appraisal

of the criteria of cerebral death. A summary state-ment. A collaboration study. JAMA 1977; 237: 982-986.

6 Shio K G, Nemoto E M, Nemmer J. Dose of thio-pental and phenytoin for maximal therapeutic effects in cerebral ischem ic anoxia. Crit Care Med1983; 2 : 452-459.

7 Woodcock J, Ropper A H, Kennedy S K. High dose barbiturates in non-traumatic brain swelling: ICP reduction and effect on outcome. Stroke 1983; 13 : 785-787.

Mohandas A, Chou S N. Brain death. A· clinical and pathological study. Journal of Neurosurgery 1971; 35: 211-217.

10 Mitchell 0 C, De La Torre E, AlexanderJ Eet.al. The

nonfilling phenomenon during angiography in acute intracranial hypertension. J Neuro Surg 1962; 19 : 766-774.

11 Brock M, Schurmann K, Hadjidimos A A. Cerebral blood flow and cerebral death. Acta Neurochir (Wien) 1969; 20 : 195-209.

12 Goodman J M, Mishkin F S, Dyken M. Determination of brain death by isotope angiography. JAMA 1969; 209 : 1869-1872.

13 Oldendorf W H. Absolute measurement of brain flow using non-diffusable isotopes. In 'International

Sym-posium on Cerebral Blood Flow. Heidelberq, 'New

York: Springer-Verlag, 1969: 53-59.

14 Braunstein P, Korein J, KricheffI.Bedside assessment of cerebral circulation. Lancet 1972; 1 : 1291-1292.

15 Braunstein P, Kricheff I, Korein J, Corey K. Cerebral Death: A rapid and reliable diagnostic adjunct using

radioisotopes. Journal of Nuclear Medicine 1972;

25 : 1047 -1 049.

16 Gomes A S, Hallinan J M. Intravenous digital sub-traction angiography in the diagnosis of brain death.

AJNR 1983;4: 21-24.

1 7 Vatne K, Nakstad P, Lundar T. Digital subtraction angiography (DSA) in the evaluation of brain death.

Neuroradiology 1985; 27 : 155-157.

18 Silverman 0, Masland R L, Saunders M G and Schuab R S. Minimal electroencephalographic recording techniques in suspected cerebral death. Electroenceph

Clin Neurophysiol 1969a; 27 : 731.

19 Boyd S G and Harden A. Nenonatal auditory brain-stem response cannot reliably diagnose brainstem death. Arch Dis Child 1985; 60 : 396.

20 Young R R. Errors in the interpretation of E~Gs in a series of 250 patients with irreversi ble coma:

Electroenceph Clin Neurophysiology 1974; 37 : 430.